About
Tanglet> * Tanglet → I can try to generate a word list using the Thai
> dictionary (I've seen there's one), but, as I strictly don't know
> anything about how Thai works, it may generate stupid results… I
> developed this strategy for alphabets I know (Latin and Cyrillic) or
> can guess (Greek). I don't know if it can be used for any other
> alphabet. We may prefer let it in English or remove it!>I understood that
> Thai works with consonants and vowels, so it might be relevant to generate
> Thai word lists >for Tanglet. However, it seems Thai words are usually short
> (1 or 2 syllabus) and many accents are used to >modify the vowel sound. So
> finally I'm not sure Tanglet is really adapted for such language.
My suggestion is to have Tanglet use the English word list, with the user
interface in Thai. It would be great to have it working in Thai, but from what
I understand of how Tanglet works it seems that, the way it is designed, it may
not work in the Thai language.
Here just two examples.1. Number of rolled out letters v/s total of alphabet
letters
A simple common word like "water" น้ำ is made of one syllable composed of: a
consonant, a vowel combination and a tone mark. Another simple common word like
"ice" is made of two syllables: น้ำแข็ง (water + hard). The word "ice" will
require seven elements available in the grid. I assume that Tanglet rolls the
letters randomly so in the 4X4 grid we may not be able to write more than two
or three words or not at all because Thai has 44 consonants, some 28 vowel
combinations, plus tone marks and some other symbols used to write words.
Tanglet will roll out 16 symbols out of more than 80. In English it rolls out
16 out of 26.2. Letter sequencing and writing rules.Each vowel has a fixed
position in writing, that is on top, at the bottom, to the left or to the right
of the consonant. A vowel combination made of two (or three) symbols will
require the consonant in the syllable (=word) to be written in a fixed
position in that combination. If we provide the double (or triple) vowel as an
alphabet letter, in Tanglet we cannot place the consonant in between the two
vowel elements. If we provide the two (or three) elements separately (as they
are in the code page table, in line D and E) and one of the two (or three)
elements is not rolled out, we can not use that vowel combination.
There are many other issues, and it would take long and a lot of work to
achieve the task. Amedeo please visit:http://www.saomaiedufund.info
On Friday, March 6, 2015 10:20 PM, "[email protected]" <[email protected]>
wrote:
Hello!
Here are some news about translation issues with Thai.
> Concerning application specific issues:
>
> * Pysycache → I think we had to force the US keyboard layout on the
> Chinese version to get it work correctly.
Hem, this was for Childsplay, not Pysycache, no?
Anyway, the trick for Childsplay has now been improved and can handle various
languages easily. I suspect this was not correctly working anymore indeed, in
DDL 2.x. So what I did is that the application executable file has been renamed
and a script launches it instead. This script has the name of the application
(it becomes then the application executable file, modified). When the script
detects the language alphabet cannot work with Childsplay, it changes the
keyboard layout to US. When the application exits, the keyboard layout is
restored.
Concerning Pysycache, I use the same trick: a script launches the real
application, taking care of adding an option to change the font whenever
required. I haven't checked whether the keyboard layout must be changed or not,
I'll check.
> * Tanglet → I can try to generate a word list using the Thai
> dictionary (I've seen there's one), but, as I strictly don't know
> anything about how Thai works, it may generate stupid results… I
> developed this strategy for alphabets I know (Latin and Cyrillic) or
> can guess (Greek). I don't know if it can be used for any other
> alphabet. We may prefer let it in English or remove it!
I understood that Thai works with consonants and vowels, so it might be
relevant to generate Thai word lists for Tanglet. However, it seems Thai words
are usually short (1 or 2 syllabus) and many accents are used to modify the
vowel sound. So finally I'm not sure Tanglet is really adapted for such
language.
> * Kgeography → many places/countries may already be translated in
> Linux global system strings. The related files are named 'iso*.mo'
> in /usr/share/locale/*/LC_MESSAGES/. That said, I've never written a
> tool to move translation from one app to another. This may be the
> opportunity to start such work!
See a previous, dedicated thread.
Cheers,
JM.
_______________________________________________
Doudoulinux-lang mailing list
[email protected]
https://mail.gna.org/listinfo/doudoulinux-lang
_______________________________________________
Doudoulinux-lang mailing list
[email protected]
https://mail.gna.org/listinfo/doudoulinux-lang
